Cable Fire Test Prediction with Small-Scale Machine Learning

Solution Overview

Problem

Conventional large-scale product tests for cables, such as EN 50399, are costly and cumbersome, discouraging manufacturers from producing innovative products due to the need for extensive cable specimens and lengthy testing processes.

Innovation Solution

A computer-implemented method using machine learning models to predict the outcome of large-scale product tests based on small-scale test results, employing multiple machine learning techniques to analyze and classify products accurately and efficiently.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If large-scale product tests (e.g., EN 50399) are performed to ensure fire safety compliance, then test reliability and measurement precision are improved, but testing cost and time increase significantly

Engineering Contradiction:

Improvefire safety compliance accuracyVSAvoidtesting time

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The patent creates a digital copy of the large-scale fire test through machine learning models. Small-scale test data is used to train ML models that simulate and predict the outcomes of large-scale EN 50399 tests, providing a virtual replica that maintains accuracy while reducing physical testing requirements

Inventive Principle:

Principle #26Copying

Solution Approach 2:

The patent replaces the physical mechanical fire testing process with a computational simulation system. Machine learning algorithms process small-scale test data and generate predictions for large-scale test outcomes, substituting physical experimentation with algorithmic modeling

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Reliability

If large-scale product tests are performed to ensure fire safety compliance, then test reliability is improved, but manufacturing cost increases

Engineering Contradiction:

Improvefire safety compliance accuracyVSAvoidmanufacturing cost

Core Design Contradiction:

ReliabilityVSEase of manufacture

Solution Approach 1:

The patent creates a digital copy of the large-scale fire test through machine learning models. Small-scale test data is used to train ML models that simulate and predict the outcomes of large-scale EN 50399 tests, providing a virtual replica that maintains accuracy while reducing physical testing requirements

Inventive Principle:

Principle #26Copying

Solution Approach 2:

The patent changes the scale parameter of the test from large-scale to small-scale. By training machine learning models on small-scale test data, the system can predict large-scale test outcomes without requiring the actual large-scale physical testing, thereby reducing material and operational costs

Inventive Principle:

Principle #35Parameter changes

3Reliability

If multiple cable specimens of sufficient length are produced for testing, then test reliability is improved, but manufacturing complexity and cost increase

Engineering Contradiction:

Improvetest result accuracyVSAvoidspecimen production complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The patent creates a digital copy of the large-scale fire test through machine learning models. Small-scale test data is used to train ML models that simulate and predict the outcomes of large-scale EN 50399 tests, providing a virtual replica that maintains accuracy while reducing physical testing requirements

Inventive Principle:

Principle #26Copying

Solution Approach 2:

The patent applies partial action by using only small-scale test data instead of requiring full large-scale test specimens. The machine learning model processes this partial data to generate predictions that would otherwise require complete large-scale testing, reducing the quantity of cable material needed

Inventive Principle:

Principle #16Partial or excessive action

Data Source

PatentEP4589494A1Machine learning technologies for predicting results of cable fire tests

Publication Date: 2025.07.23 UL LLC

Figure 1A
Figure 1B
Figure 2

AI summary

Systems and methods for using machine learning models to predict an outcome of a product test are described. According to certain aspects, an electronic device may calculate, based on a received set of small-scale results as a first input to a first machine learning model of a plurality of machine learning models, a first result predicting an outcome of the product tested according to the large-scale product test. The electronic device may then calculate, based on the set of small-scale results as a second input to at least one second machine learning model of the plurality of machine learning models, a second result predicting the outcome of the product tested according to the large-scale product test. The electronic device may then predict an outcome of the large-scale product test based at least on the first result and the second result.